Aspects of reproductive biology of the scalloped hammerhead shark, Sphyrna lewini, off northeastern Brazil

Page created by Regina Clarke
 
CONTINUE READING
Environmental Biology of Fishes 61: 151–159, 2001.
© 2001 Kluwer Academic Publishers. Printed in the Netherlands.

Aspects of reproductive biology of the scalloped hammerhead shark,
Sphyrna lewini, off northeastern Brazil

Fabio Hazin, Alessandra Fischer & Matt Broadhursta
Universidade Federal Rural de Pernambuco, Departamento de Pesca, Laboratório de Oceanografia Pesqueira,
Av. Dom Manuel de Medeiros, s/n, Dois Irmãos, Recife-PE, Brasil (e-mail: fhvhazin@elogica.com.br)
a
  Corresponding author

Received 3 March 2000        Accepted 30 November 2000

Key words: maturation, life history, size at maturity, life cycle

Synopsis

Ninety four scalloped hammerhead sharks, Sphyrna lewini (53 females and 41 males) ranging in size from 121 to
321 cm total length (TL), were collected from surface gillnetters operating off northeastern Brazil and throughout
the southwestern equatorial Atlantic Ocean between January and December 1996. A common regression for TL
and eviscerated weight (EW) was calculated as, log EW = −11.786 + 2.889 log TL. Females and males were
categorised into reproductive stages (4 and 2, respectively) according to morphological changes in their gonads.
Size at sexual maturity for females was estimated to be 240 cm, while males appeared to mature at between 180
and 200 cm. Gravid females had between 2 and 21 embryos or pups, varying in TL from 3 to 38 cm. There was
no relationship between maternal length and size of litter. Copulation and parturition appear to occur outside the
sampled area and possibly closer to the coast. With the exception of slightly lower uterine and ovarian fecundities,
the results support the few existing data on the reproductive cycle of S. lewini in other areas.

Introduction                                                     (Clarke 1971), Southern Africa (Bass et al. 1975), the
                                                                 Gulf of Mexico (Branstetter 1981, 1987), Taiwan (Chen
The scalloped hammerhead shark, Sphyrna lewini, is               et al. 1988), northern Australia (Stevens & Lyle 1989),
a semi-oceanic species globally distributed through-             northern Brazil (Lessa et al. 1998) and Senegal (Capapé
out tropical and temperate oceans (Compagno 1984).               et al. 1998). Despite this research, with the exception of
Characterised by a range of habitats, the young typi-            the work done by Chen et al. (1988) off Taiwan, there
cally occur in shallow bays and estuaries, while adults          is little information available on females and particu-
tend to frequent continental shelves and adjacent waters         larly gravid individuals throughout their distribution.
to depths of 275 m (Clarke 1971, Compagno 1984,                  This is attributed to sexual segregation of adults and
Klimley 1987). The abundance and accessibility of                the preference of females for deeper areas away from
S. lewini in these areas means that it often comprises           inshore fishing grounds (Branstetter 1987, Stevens &
a large proportion of elasmobranch catches from com-             Lyle 1989).
mercial fishing activities, such as shrimp trawling (e.g.           Until recently, such disequilibrium in catch compo-
Castro 1993), longlining (e.g. Branstetter 1987, Chen            sitions of S. lewini precluded descriptions of the repro-
et al. 1988) and gillnetting (e.g. Stevens & Lyle 1989).         ductive biology of females off northeastern Brazil.
   Owing to availability and commercial impor-                   Artisanal gillnetters working in and around estuaries
tance, aspects of their reproductive biology have                along the north coast frequently catch juveniles and
been described at various locations including, Hawaii            some information was presented on their biology in
152

a previous study (Lessa et al. 1998). While adjacent        eviscerated weight ×1000. Insufficient data precluded
oceanic areas, including various sea mounts and             the analysis of GSI per month and data were pooled
offshore islands, have been extensively fished by a         across seasons throughout the year. There were insuffi-
pelagic longline fleet for the past 20 years (see Hazin     cient mature females caught across seasons to provide
et al. 1998 for details) catches of S. lewini have always   meaningful analysis of temporal changes in their GSI.
been very low (less than 0.05% of total shark catch –          Inferences on stages of maturation were made
Hazin et al. 1990). In 1996, however, several vessels in    according to definitions provided in previous stud-
this fleet employed monofilament gillnets (for a period     ies (e.g. Springer 1960, Pratt 1979, Branstetter 1981).
of 12 months) and caught relatively greater numbers         Depending on the development of the oviducal gland,
of adult S. lewini (approx. 13% of total catch) in the      uteri, MOFD and ovary, females were separated into
same areas fished with longlines. Given the absence of      4 stages. Specimens were considered juvenile if they
information on reproductive biology of adults in the        had undeveloped sexual organs, filiform uteri, and no
southwestern Atlantic Ocean and the paucity of data         vitellogenic activity in their ovaries. Maturing females
on females from most areas throughout the world, our        had enlarged oviducal glands and showed evidence
aim in the present study was to examine some of the         of vitellogenesis. Ovulating females had uterine eggs
specimens caught to supplement the existing literature.     and mature oocytes in their ovaries, while in gravid
                                                            specimens ovulation was complete. Male maturation
                                                            was evaluated according to development of the testes
Material and methods                                        and calcification of claspers. Individuals with relatively
                                                            short, flexible claspers and filiform ampullae ductus
This work was based on 94 sharks (53 females and            deferens were considered juvenile. Adults were char-
41 males) collected from the catches of commercial          acterised by elongated and calcified claspers.
fishing vessels using drifting gillnets in the southwest-
ern equatorial Atlantic (>150 km from the coast of
northeastern Brazil, Figure 1) between January and          Statistical analyses
December 1996. The gillnets used were up to 12 m in
depth, had a buoyed headline of between 1 and 7 km          Linear regressions of EW and TL were calculated sep-
in length, and consisted of monofilament mesh with a        arately for males and females and then compared using
size between the knots (stretched mesh) of 17 to 30 cm.     appropriate analysis of co-variance (ANCOVA). Size-
Nets were set by hand at the surface each day at approx.    frequency distributions of males and females were
18:00 h and hauled the following morning at approx.         compared using a two-sample Kolmogorov–Smirnov
5:00 h.                                                     test (p = 0.05). Regression analysis was used to exam-
   Immediately after being landed, sharks were mea-         ine the relationships between litter size and TL for
sured (total length – TL and fork length – FL, see          mature females. Data describing seasonal changes in
Garrick 1982 for details), dissected and weighted           GSI for mature males were tested for heteroscedasticity
(eviscerated weight – EW). Reproductive organs were         using Bartlett’s test, and then analysed in an unbalanced
removed and stored in a solution of 10% formalin in         one-factor analysis of variance (ANOVA). Chi-squared
sea water prior to being transported to the laboratory.     goodness-of-fit tests were used to examine the hypoth-
Data collected from females included weight of the          esis of an equal sex ratio among numbers of S. lewini
functional right ovary, maximum ovarian follicle diam-      sampled and pups from gravid females.
eter (MOFD), width of the oviducal gland and largest
uterus and if present, the number of eggs or embryos
and the length and sex of pups. Using the methods           Results
described by Pratt (1993), the oviducal glands of 36
females were examined for the presence of spermato-         Although there was some temporal variability between
zoa. The weight of testes, length and calcification state   sexes caught during the period examined (Table 1),
of claspers and the presence of seminal fluid in ampulla    the pooled sex ratio of males to females (1 : 1.29)
ductus deferens were recorded from males. Reproduc-         was not significantly different from 1 : 1 (χ 2 = 1.53,
tive organs were measured to the nearest 0.1 mm using       p > 0.05). The majority of specimens were sampled in
vernier calipers. A gonosomatic index (GSI) for mature      the latter half of the year and mostly between July and
males was calculated by dividing weight of testes by        November. The Kolmogorov–Smirnov test detected
153

                                           Figure 1. Location of fishing area.

significant differences in size-frequency compositions          n = 36) and females (log EW = −12.032 + 2.935
between males (ranging from 123 to 321 cm TL) and               log TL, r2 = 0.933, n = 50). A common regression
females (ranging from 121 to 273 cm TL), with pro-              was calculated as log EW = −11.786 + 2.889 log TL.
portionally more females between 201 and 279 cm TL
captured. One male, 321 cm TL, represents the largest           Females
measured specimen to date. ANCOVA failed to detect
significant differences in regression coefficients (F =         Nine of the specimens sampled were juveniles, rang-
0.017, p > 0.05) or elevations (F = 0.3707, p > 0.05)           ing in size from 121 to 220 cm TL. They had light
between the regressions of log EW and log TL for                ovaries, narrow oviducal glands and uteri and unde-
males (log EW = −11.883+2.908 log TL, r2 = 0.962,               veloped ovarian follicles (Table 2, Figures 2a–c).
154

Table 1. Numbers of S. lewini sampled during each month in              some evidence to suggest that they were ready for
1996.                                                                   ovulation shortly after parturition. Spermatozoa were
Month of         Males                  Females                 Total
                                                                        detected in only one of the females examined: a 246 cm
capture                                                                 TL gravid individual. None of the females showed any
                 Juveniles   Adults     Juveniles     Adults
                                                                        evidence of mating scars.
January          1            4          5            0         10         The uterine contents of the 10 gravid females cap-
February         0            0          0            0          0      tured during the study are summarised in Table 3.
March            0            3          1            0          4
April            0            1          0            0          1
                                                                        Litter size ranged from 2 to 21 embryos and pups
May              0            0          0            0          0      with a mean (SE) of 14.3 (2.31). All embryos from
June             2            6          0            0          8      individual gravid females showed a similar state of
July             1           13          3            2         19      development. They were longitudinally oriented in
August           0            3         12            3         18      individual chambers and in the same direction as
September        0            1          0            0          1      the mother. The sex ratios of embryos in individual
October          1            2          8            8         19
November         0            3          3            6         12
                                                                        females showed considerable variation, although the
December         0            0          1            1          2      total pooled ratio (1 : 1.74) was significantly biased
                                                                        towards females (χ 2 = 8.67, p < 0.01). No signifi-
                                                                        cant linear relationship was detected between size of
                                                                        gravid female and number of embryos (F = 0.034,
Table 2. Characteristics of female S. lewini in each maturation         p > 0.05).
stage and the length range and number of specimens. All lengths
and widths are in centimeters and weights in grams.
                                                                        Males
Characteristic       Juvenile     Maturing    Ovulating Gravid
Width of
155

Figure 2. Relationships between total length and: a – width of oviducal gland, b – width of uterus, c – MOFD, and d – weight of ovary
across the various stages of maturation for females.
156

Iwamoto 1981, Chen et al. 1988. The available data                       greater numbers of females (i.e. ratio of 1 : 1.29) were
support these trends off northern Brazil, since Lessa                    sampled in the present study.
et al. (1998) recorded larger and more juveniles males                      Evaluation of the maturation stages of female
(46 to 173 cm TL) than females (45 to 149 cm TL) in                      S. lewini showed considerable overlap between cohorts
shallow coastal areas. Although not significant, slightly                (Table 2, Figure 2), however, the minimum sizes at
                                                                         each stage are comparable to observations by Stevens
                                                                         & Lyle (1987) who noted TLs of 228, 229 and 238 cm
 Table 3. Summaries of the month of capture, total length (TL),          for mature non-virgin, pre-ovulatory, and pregnant
 and number and sex of embryos found in the 10 gravid S. lewini          females, respectively. Because many of the individuals
 (TL in cm, ∗ = not sexed).
                                                                         greater than 240 cm in the present study had enlarged
 Month of     TL of gravid TL range   Number of embryos                  oviducal glands, thick uteri and heavy ovaries (Figure
 capture      specimen     of embryos Male Female Total                  2), it is likely sexual maturity was approached at this
                                           ∗       ∗
                                                                         length. This estimate is larger than the sizes of 210 and
 July         258.9            3 – 5.5                     14
 August       246              3– 4        ∗       ∗
                                                           11
                                                                         200 cm TL proposed by Chen et al. (1988) and Stevens
 August       254            17 – 20       9     11        20            & Lyle (1989) for specimens captured in the western
 October      253.3         10.5 – 13      6     12        18            Pacific Ocean, but similar to that (i.e. 250 cm TL) pro-
 October      247.2         11.5 – 13.5    0      2         2            vided by Branstetter (1987) for individuals from the
 October      272.8          13 – 15.5     4     10        14            Gulf of Mexico (Table 4).
 October      255.1          14 – 17       2      0         2               In contrast to females, males appeared to mature at
 October      249.9         29.9 – 33.7   11     10        21
                                                                         a shorter TL, although the few small specimens cap-
 November     259           32.5 – 35.5    9     12        21
 November     244            32 – 38       2     18        20            tured mean it is difficult to provide an accurate esti-
                                                                         mate of the size range. Three of the 5 males classed

      Figure 3. Relationships between total length and: a – length of clasper, and b – weight of testes for juvenile and adult males.
157

as juveniles (176, 180 and 195 cm TL), had calcifying                  variability among estimates of size at maturity, ranging
claspers (i.e. semi-rigid) that were considerably longer               from 140 to 198 cm TL (Table 4).
than those in the remaining two juveniles (123 and                        Aspects of the reproductive cycle of S. lewini have
159 cm TL – Figure 3a). Given these observations and                   been proposed by Chen et al. (1988) and collaborated
that all specimens longer than 200 cm TL were adults,                  by data collected on individual females in other stud-
males probably approached maturity at between 180                      ies (Table 4). Vitellogenesis and gestation each appear
and 200 cm. In the only other relevant study off north-                to take about 10 months with ovulation and parturition
ern Brazil, Lessa et al. (1998) observed 53 males (46                  mostly occurring between July and October (autumn)
to 173 cm TL) and found three (94, 150 and 173 cm                      and May and July (summer), respectively, in the north-
TL) with evidence of calcified claspers (although the                  ern hemisphere. The data recorded from the 10 preg-
length of the clasper in the shortest specimen was                     nant females examined in the present study also support
less than 7 cm). Other studies have shown considerable                 this cycle. For example, the smallest embryos recorded
                                                                       were between 3 and 5.5 cm TL from a gravid female
                                                                       caught in July, while the largest (32 to 38 cm TL) were
                                                                       recorded from two females in November (Table 3).
                                                                       Gestation probably starts in autumn, and given that the
                                                                       largest observed embryos were shorter than the 45 to
                                                                       50 cm size at birth postulated in most studies (Table 4)
                                                                       parturition likely occurs during mid to late summer.
                                                                       The lack of near-term females in the sample, combined
                                                                       with the observation by Lessa et al. (1998) of several
                                                                       neonates (>45 cm TL) in estuarine areas off northern
                                                                       Brazil provides some evidence to indicate that partu-
                                                                       rition occurs outside the examined area, and probably
                                                                       closer to the coast.
                                                                          Some evidence of the simultaneous development of
                                                                       ovarian follicles and embryos, along with the presence
                                                                       of spermatozoa in one gravid female, suggests that
Figure 4. Differences in mean GSI (± se) of mature males               females may be capable of giving birth annually (see
between seasons.

Table 4. Summary of studies describing some aspects of the reproductive biology of S. lewini. TL in cm (Na = not available).

Sampled         TL at maturity        Gestation period Parturition           Litter size TL at birth Reference
location        Males      Females
Hawaii           Na        Na         Na                Year round            15–31       40–50      Clarke (1971)
                                                         (peak in summer)
Southern      140–165      212        Na                Summer                30          50         Bass et al. (1975)
 Mozambique
Gulf of       180          250        12 months         Na                   > 30         49         Branstetter (1987)
 Mexico
Northeastern  198          210        10 months         Summer                12–38     >47          Chen et al. (1988)
 Taiwan
Northern      140–160      200        9–10 months       Summer                13–23       45–50      Stevens & Lyle (1989)
 Australia
Senegal       Na           Na         Na                Summer                18–22       37–52      Capapé et al. (1998)
Northern     38          Present study
 Brazil
158

also Chen et al. 1988, Capapé et al. 1998). In partial        Acknowledgements
support of this, adult males showed large variation in
their weight of testes (Figure 3b) which typically is          This study was funded by the Comissão Interministerial
associated with a seasonal fluctuation in sperm produc-        para os Recursos do Mar (CIRM) through the Programa
tion (e.g. Hazin et al. 2000). Further, there was some         Nacional de Avaliação dos Recursos Vivos da Zona
evidence of a peak in mean GSI of mature males during          Econômica Exclusiva (REVIZEE) and the Conselho
autumn (Figure 4) which may coincide with a mating             Nacional de Ensino e Pesquisa (CNPq). Thanks are
season, however ANOVA failed to detect any signifi-            extended to Norte Pesca for their assistance in compil-
cant temporal differences in mean GSI. A smaller stan-         ing the catch data and to R. Lessa for helpful comments
dard error, achieved through greater replication (e.g.         and advice.
only two males were caught during autumn) would
be required to detect clear patterns of difference. The
absence of additional data precludes any accurate deter-       References cited
mination of the time or location of copulation, although
because none of the sampled females showed evidence            Bass, A. J., J.D. D’Aubrey & N. Kistnasamy. 1975. Sharks of the
of fresh mating scars (normally associated with this             east coast of southern Africa. III. The families Carcharinidae
                                                                 (excluding Mustelus and Carcharhinus) and Sphyrnidae.
species – Wakabayashi & Iwamoto 1981) this proba-                Invest. Rep. (S. Afr. Ass. mar. Biol. Res.) 38: 1–100.
bly occurs outside the sampled area.                           Bigelow, H.B. & W.C. Schroeder. 1948. Sharks. pp. 59–546.
   Unlike the results from Chen et al. (1988), we failed         In: Fishes of the Western North Atlantic, Pt. 1, Lancelets,
to detect any relationship between maternal length and           Cyclostomes, and Sharks, Mem. Sears Found. Mar. Res., New
size of litter for the 10 females examined. Further, the         Haven.
uterine fecundity (between 2 and 21 pups – Table 3)            Branstetter, S. 1981. Biological notes on the sharks of the north
                                                                 central Gulf of Mexico. Cont. Mar. Sci. 24: 13–34.
of these females was considerably lower than that (e.g.        Branstetter, S. 1987. Age, growth and reproductive biology of
between 12 and 38 pups) recorded in previous work                the silky shark, Carcharhinus falciformis, and the scalloped
(Table 4). This later result may be biased by abor-              hammerhead, Sphyrna lewini, from the northwestern Gulf of
tion during capture in the gillnets (owing to trauma             Mexico. Env. Biol. Fish. 19: 161–173.
incurred as sharks struggle to free themselves – e.g.          Capapé, C., M. Diop & M. N’Dao. 1998. Record of four preg-
                                                                 nant females of the scalloped hammerhead, Sphyrna lewini
Clarke 1971), however, ovarian fecundity in maturing
                                                                 (Sphyrnidae) in Senegalese waters. Cybium 22: 89–93.
specimens was also generally lower (i.e. between 7 to          Castro, J.I. 1993. The shark nursery of Bulls Bay, South Carolina,
40 oocytes) than previous observations (e.g. 37 to 46            with a review of the shark nurseries of the southeastern coast
oocytes – Capapé et al. 1988). It is difficult to postulate     of the United States. Env. Biol. Fish. 38: 37–48.
reasons for such relatively low fecundity, other than          Chen, C., T. Leu & S. Joung. 1988. Notes on reproduction in the
that it may be an inherent characteristic of the stock           scalloped hammerhead, Sphyrna lewini in northeastern Taiwan
                                                                 waters. U. S. Fish. Bull. 86: 389–393.
off northeastern Brazil. More detailed analysis, involv-       Clark, E. & K. von Schmidt. 1965. Sharks of the central gulf coast
ing a greater sample size, would be required to validate         of Florida. Bull. Mar. Sci. 15: 13–83.
this observation. The significantly greater number of          Clarke, T.A. 1971. The ecology of the scalloped hammerhead
female pups in litters is similar to observations made           shark, Sphyrna lewini, in Hawaii. Pac. Sci. 25: 133–144.
by Capapé et al. (1998), but differs from the equal pro-      Compagno, L.V.J. 1984. FAO species catalogue, Vol 4, Sharks
portions recorded by Chen et al. (1988) and Stevens              of the world. An annotated and illustrated catalogue of shark
                                                                 species known to date. Part 2, Carcharhiniformes. FAO Fish.
& Lyle (1987).                                                   Synop. 125, Vol 4, Pt. 2. 655 pp.
   With the exception of some anomalies concerning             Garrick, J.A.F. 1982. Sharks of the genus Carcharhinus. NOAA
fecundity, the results presented here concur with much           Tech. Rep. NMFS Circ. 445. 193 pp.
of the existing information available on the reproduc-         Hazin, F.H.V., A.A. Couto, K. Kihara, K. Otsuka & M.
tive biology of S. lewini. Further fishery-independent           Ishino. 1990. Distribution and abundance of pelagic sharks
research is still required off northeastern Brazil, how-         in the south-western equatorial Atlantic. J. Tokyo Univ. Fish.
                                                                 77: 51–64.
ever, to locate areas of copulation and parturition            Hazin, F.H.V., F.M. Lucena, T.S.A.L. Souza, C.E. Boeckman,
and provide some analysis of other stock-specific                M.K. Broadhurst & R.C. Menni. 2000. Maturation of the night
life history parameters, as well as abundances and               shark, Carcharhinus signatus, in the south-western equatorial
distributions.                                                   Atlantic Ocean. Bull. Mar. Sci. 66: 173–185.
159

Hazin, F.H.V., J.R. Zagaglia, M.K. Broadhurst, P.E.P. Travas-        Pratt, H.L. 1993. The storage of spermatozoa in the oviducal
  sos & T.R.Q. Bezerra. 1998. Review of a small-scale pelagic          glands of western North Atlantic sharks. Env. Biol. Fish. 38:
  longline fishery off northeastern Brazil. Mar. Fish. Rev. 60:        139–149.
  1–8.                                                               Springer, S. 1960. Natural history of the sandbar shark, Eulamia
Klimley, A.P. 1987. The determinants of sexual segregation in the      milberti. U. S. Fish. Bull. 61: 1–38.
  scalloped hammerhead shark, Sphyrna lewini. Env. Biol. Fish.       Stevens, J.D. & J.M. Lyle. 1989. Biology of three hammerhead
  18: 27–40.                                                           sharks (Eusphyra blochii, Sphyrna mokarran and S. lewini)
Lessa, R., R.C. Menni & F. Lucena. 1998. Biological observations       from northern Australia. Aust. J. Mar. Freshwater Res. 40:
  on Sphyrna lewini and S. tudes (Chondrichthyes, Sphyrnidae)          129–146.
  from northern Brazil. Vie Milieu 48: 203–213.                      Wakabayashi, K. & T. Iwamoto. 1981. Schooling of the scalloped
Pratt, H.L. 1979. Reproduction in the blue shark, Prionace glauca.     hammerhead shark, Sphyrna lewini, in the Gulf of California.
  U. S. Fish. Bull. 77: 445–470.                                       U. S. Fish. Bull. 79: 356–359.
You can also read